An Analytical Generator for Wave Energy Converter Optimization Using Co-Design Methods

In numerical optimization a common approach is to sweep over a continuous parameter space. Within wave energy converter optimization, WecOptTool recommends the same approach to identify trends. WecOptTool is an open-source software that uses co-design methods that couple control design with mechanical design due to the highly dependent nature between the two. Although WecOptTool is quite useful, its optimized results are dynamic properties of an idealized system which are one step removed from the actual mechanical components necessary to achieve its results. Therefore, the physical components necessary to match these optimized dynamic outputs might not actually exist off the shelf.

Our previous work in “Preliminary Co-Design and Structured Innovation” from UMERC 2023, identified two areas for WEC innovation- reactive mechanical components as well as high generator impact on power performance through dynamics and cost. Using our knowledge from the latter finding, the discontinuity between an economically attainable or unattainable design could be due to commercially available generator technical specifications. Today, commercially available generators may have a gap in which wave energy converter-specific generators could be designed. A deep dive into commercially available off the shelf generator component options and how an idealized generator compares in terms of dynamics and cost is explored in this study to develop a power and cost-effective design.

In the currently published PTO models of WecOptTool, the free variables are relatively independent. We create a more dependent and physically realistic model by breaking down the variables into first principles such as generator windings and material cost. A sensitivity analysis of power and cost is conducted with respect to the generator parameters. Additionally, we use WecOptTool to optimize a generator within the RM3 which is subject to generator torque constraints, WEC stroke constraints, and other force constraints. To develop a power and cost-effective design, we develop analytical models for both the generator dynamics and generator cost and tune the cost model with existing catalog data. The idealized generator can be placed alongside commercially available off the shelf (COTS) generator catalog data, or it may exist far outside of the COTS data range. One impactful outcome of this work may identify a gap where generators used in wave energy converter design could be created. For example, most generators are commonly used in applications where minimizing moment of inertia and delivering a reasonable torque for the application is desired. However, in the case of wave energy converters, minimized moment of inertia may not be as important as achieving a higher torque constant to optimize power output. Another outcome may be useful to developers looking for commercially available off the shelf generators that are, or are close to, to the optimized solution from our study.